A remarkable variety of clinical variables have been shown to be related to the body fat distribution. Many studies have shown that the waist hip ratio (WHR) is not only a simple index of the fat distribution pattern, but is also a powerful predictor of risk factors, diseases, and mortality. Despite extensive literature, systematic information on the distribution of values for the two components of the WHR (waist and hip circumference), for the ratio itself, and for changes in these values with change of weight is not available. The effects of age, sex and change in weight on WHR have been examined in the BLSA subjects. The effects of age, WHR body fatness, and physical activity and fitness levels have been examined with respect to their contribution to the variance in glucose tolerance. In step-wise multiple regression analyses, age remains as an independent significant correlate fatness and WHR also enter as significant determinants. Physical activity level and physical fitness make very little contribution in either men or women, but may still exert an important effect by influencing body fatness. An analysis of the effects of cigarette smoking on the pattern of distribution of body fat has produced surprising results. In BLSA men, cigarette smokers (S) were leaner than non- smokers (non-S) except for the heaviest smokers. Despite the generally lower body weights in smokers, the WHR rose progressively with the number of cigarettes smoked. Furthermore, a change in smoking habit (S to non-S or non-S to S) gave unexpected results when the changes in body weight were considered. Despite the gain in weight in the "quitters" (S to non-S group), WHR did not increase; in the "starters" (non-S to S), despite weight loss, the WHR actually increased. Thus, the decision to initiate or to continue smoking in order to control body weight is doubly unwise from the health standpoint. The direct harmful effects of smoking will be reinforced by a more dangerous pattern of fat distribution. Furthermore, smokers who quit can be reassured that their weight gain (not an obligatory result of course) will at least not cause any worsening of their fat distribution pattern.